Identification of cancer drug targets using high throughput screens of tumor cell lines has led to a number of agents presently in clinical trials. In addition, recent advances in drugs that attack immune cells within tumors, such as ?CTLA4 and ?PD-1, have highlighted the importance of immune modulation as a strategy for cancer therapy. The next phase of cancer drug target discovery will seek to integrate these strategies to identify combinations of drugs that most efficiently target both tumor cells and the immune components in advanced cancers. The goal of this proposal is to identify and validate these combinations using large-scale data mining and mouse pre-clinical cancer models that mimic the major genetic features of human cancer. This proposal addresses both mechanisms of immune escape by a) finding genetic targets that may enhance tumor mutation load, and b) carrying out high throughout screens in T cells or myeloid cells for targets that promote immune cell infiltration. We will exploit unique mouse models that mirror major genetic categories of human cancer ? high vs low mutation load, and strong vs weak immune infiltrate. Applying single-cell RNAseq and mass cytometric proteomic analyses, cutting edge immune composition databases and novel computational network approaches to cancer target discovery using existing large databases, we propose to identify vulnerabilities addressed by combining small molecule drugs with immunotherapy. We will make immunologically ?cold? tumors, that do not engage the immune system, into ?hot? tumors that present more or stronger antigens, or that encourage infiltration by immune effector cells. To achieve this goal, we propose three highly innovative aims centered on perturbation of specific targets: first by a CRISP/Cas9 screen in immune cells of the tumor microenvironment, second through increasing antigen load in tumors to optimize immune recognition and finally through a network-based identification of tumor-expressed targets that may confer susceptibility to existing immune-oncology therapies. This represents a true `network' of our collective expertise as well as a measured collection of candidate and screening approaches. AIM 1 ?We will perform CRISPR screens in monocytes and T-cells to identify genes associated with tumor entry and function in two distinct tumor types. AIM 2? We will use genetic or pharmacological perturbation of newly generated candidate genes involved in metabolic stress and ROS-induced DNA damage to increase mutation load and antigen abundance in a tumor- specific manner, leading to improved responses to immunotherapy. AIM 3 ? We will exploit gene expression networks to identify druggable targets and pathways that augment immune responses. This proposal identifies pathways and perturbants for accelerating immunotherapies.